How Adobe Data Helped Combat Counterfeiting and Brand Abuse with Advanced Data Solutions

Counterfeiting, web spoofing, fake listings, and product replicas have surged significantly in recent years, adversely affecting the sales and growth of major global brands. Identifying these online fraudsters across millions of websites, marketplaces, and web portals is akin to searching for a needle in a haystack.

Our client, a leading revenue recovery company, leverages proprietary AI-powered software to address this challenge. Their advanced platform scans the web on behalf of businesses to uncover intellectual property (IP) infringements. This technology enables the detection of copyright violations, brand impersonations, product piracy, counterfeit goods, and distribution abuse. By providing this specialized service, the client supports over a thousand online brands, helping them monitor fraud, gather evidence, and take the necessary legal action to reclaim lost revenue.

Adobe Data played a pivotal role in enhancing the client’s AI-driven brand protection platform by delivering comprehensive data support solutions to close critical gaps.

Managing the vast influx of data scraped by their AI system for numerous brands had become increasingly challenging for the client. To alleviate this burden, they decided to outsource certain data-specific operations.

Through our discussions, we identified several key issues requiring immediate attention:

1. Data Cleaning: For the AI platform to operate efficiently, it required a steady stream of high-quality data, emphasizing the importance of regular data cleaning and hygiene monitoring.
2. Data Validation: The scraped data, primarily consisting of links to counterfeit products, needed systematic verification to ensure accuracy and reliability.
3. Data Appending: In cases where data crawling and extraction methods fell short, the collected data often lacked completeness. This required manual efforts to append missing information.
4. Web Research: When automated scraping failed to identify online scammers, manual web research became essential to uncover fraudulent activities.
5. Dedicated Data Management Team: To streamline the entire process, we proposed assembling a dedicated team of data researchers. This team would build expertise over time to refine and perfect the data management workflow.

By addressing these challenges, we aimed to enhance the efficiency and effectiveness of their AI-driven brand protection software.

“During our discussions with the team, we anticipated some of the issues that emerged. However, when we saw the full scope after our initial talks, we were both shocked and frustrated by how much of our time was being consumed addressing the fallout from problems we hadn’t even recognized as significant.”

-A business user reflecting on the requirement discussion phase

The client serves over a thousand brands, and we were initially tasked with managing only a select few.

By the end of the first month, the number of brands assigned to us had doubled. Our dedicated team’s efforts led to a 70% reduction in client involvement for feedback and quality analysis. Additionally, we enhanced accuracy rates from 72% to an impressive 95%.

For this project, our team demonstrated remarkable resilience, experimenting with various approaches, reevaluating and refining processes, and establishing quality assurance systems, all while adapting to the client’s evolving needs and workload.

Initially, we organized our team by assigning specific brands to each member of the manual data detection team. However, this approach led to quality issues and increased errors. To address this, we restructured our teams based on processes, enabling more streamlined management and improved outcomes.

Through ongoing process optimization, resource upskilling, and strict adherence to the client’s data standards, we earned their trust and became their ‘Vendor of Choice.’ This partnership resulted in several notable achievements:

• The client reported a 30% reduction in delivery timelines for their clientele.
• A long-term contract was signed, solidifying our continued partnership.
• We now manage over 50% of the client’s workload, supporting nearly half of the brands using their AI system.
• The project team grew from 6 to 101 members (and continues to expand) within just six months.

These milestones highlight the success of our collaborative efforts and our commitment to excellence.

“While there is still work to be done, we have successfully helped the client achieve a critical goal—enhancing the accuracy of their AI platform and delivering on their commitments to their clientele more effectively.”

The objective of any AI-driven tech platform is to achieve maximum precision. However, this is challenging across different websites, as each one typically has its own unique underlying patterns. Additionally, website structures evolve over time, requiring AI solutions to be adaptable. Automated crawling methods, no matter how well-trained, cannot function independently; they must operate alongside a human-in-the-loop approach to achieve the best results.

Adobe fully understands the power of this man-plus-machine approach and continues to successfully support several AI-based tech platforms with their diverse data management needs.

“As a back-office data support partner, Adobe has been an invaluable collaborator. Thanks to the unwavering support from their team and our collective efforts, we were able to enhance the efficiency of our AI-powered brand protection software, saving millions in revenue for our clients. Additionally, with dedicated resources and prompt data support, we successfully reduced our delivery timeline by 30%. We are excited about continuing our long-term partnership with their team.”

– CEO, Client Company